MRCetAKI

DrGuillaumeClaisseNéphrologie– CHUSaint-Etienne

Nature Reviews | Nephrology

180907

Days post injury

2(48h)

0

Injury

AKI AKD CKD

excluded as this approach is typically used in austere conditions and special circumstances (for example, in small children).

Persistent AKITransient versus persistent AKIVarious studies, generally limited by the patient popu-lations selected and the use of serum creatinine

changes to assess renal function, have applied different thresholds for the duration of AKI episodes and func-tional renal recovery to discriminate transient from persistent AKI (see Supplementary information S1 (table)). Regardless of disease severity, these studies demonstrate that complete and sustained reversal of an AKI episode within 48–72 h of its onset is associated with better outcomes than longer durations of AKI10–15. Based on the available data and expert opinion, the workgroup proposes using 48 h to define rapid reversal of AKI (BOX 1). The rationale for selecting 48 h rather than 72 h to define rapid reversal is to better identify high-risk patients for whom additional workup and evaluation might be warranted. Although previous stud-ies have relied primarily on serum creatinine to identify AKI, the workgroup recommends also using urine out-put criteria as recommended by KDIGO5. The impor-tance of urine output criteria in defining persistent AKI was confirmed in a 2015 study of 32,045 critically ill patients, which demonstrated that short-term and long-term risk of death or RRT is greatest for patients who meet both the serum creatinine and urine output criteria for AKI and experience these abnormalities for longer than 3 days12.

For AKI that has reversed, it is unknown when sus-tained reversal can be considered to have occurred. Although the duration of sustained reversal might be different for rapidly reversing and persistent AKI we propose a minimum of 48 h as being necessary to separate two distinct episodes of AKI. After sustained reversal has occurred, a second episode of AKI should be considered independently of the first, with new inves-tigations to exclude potentially new reversible causes or

Figure 2 | The continuum of acute kidney injury (AKI), acute kidney disease (AKD) and chronic kidney disease (CKD). AKI, AKD and CKD can form a continuum whereby initial kidney injury can lead to persistent renal injury, eventually leading to CKD. AKI is defined as an abrupt decrease in kidney function occurring over 7 days or less, whereas CKD is defined by the persistence of kidney disease for a period of >90 days. AKD describes acute or subacute damage and/or loss of kidney function for a duration of between 7 and 90 days after exposure to an AKI initiating event. Recovery from AKI within 48 h of the initiating event typically heralds rapid reversal of AKI. For patients with pre-existing CKD, the AKI event can be superimposed on CKD, with AKD existing on a background of CKD. Patients who suffer AKD with pre-existing CKD are probably at high-risk of kidney disease progression. Modified from Acute Dialysis Quality Initiative 16; www.adqi.org.

Box 1 | Definitions of AKI and AKD, initial management of AKI, and assessment of kidney function

Consensus statement 1A:Persistent acute kidney injury (AKI) is characterized by the continuance of AKI by serum creatinine or urine output criteria (as defined by KDIGO) beyond 48 h from AKI onset. Complete reversal of AKI by KDIGO criteria within 48 h of AKI onset characterizes rapid reversal of AKI (evidence grade: level 5).

Consensus statement 1B:Although the optimal duration of sustained AKI reversal is unknown, a minimum of 48 h is necessary to separate two distinct AKI episodes (evidence grade: level 5).

Consensus statement 1C:AKI and acute kidney disease (AKD) are a continuum, and persistent AKI frequently becomes AKD, defined as a condition wherein criteria for AKI stage 1 or greater persists ≥7 days after an exposure (FIG. 2; TABLE 1; evidence grade: level 4).

Consensus statement 1D:Initial management of persistent AKI should include reassessment of the underlying aetiology of AKI and precise measurement of kidney function. When persistent AKI is diagnosed, additional monitoring should be considered to re-evaluate haemodynamic and volume status, adequacy of kidney perfusion, and to identify complications of AKI, such as fluid overload, acidosis and hyperkalaemia, as these could indicate a need for renal replacement therapy. Nephrology consultation should be considered if the aetiology of AKI is not clear or subspecialist care is needed (evidence grade: level 5).

Consensus statement 1E:An urgent need exists for clinical tools to enable the precise measurement of kidney function in the setting of AKI as existing tools are impractical for routine clinical use. At present, timed urine creatinine clearance is the best available estimate of kidney function for patients with persistent AKI in the steady state (evidence grade: level 4).

Consensus statement 1F:Equations to estimate glomerular filtration rate in the setting of chronic kidney disease are not accurate for the assessment of renal function in persistent AKI (evidence grade: level 4).

C O N S E N S U S S TAT E M E N T

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 3

Chawla,NatRev,2017

Section 2: AKI DefinitionKidney International Supplements (2012) 2, 19–36; doi:10.1038/kisup.2011.32

Chapter 2.1: Definition and classification of AKI

INTRODUCTIONAKI is one of a number of conditions that affect kidneystructure and function. AKI is defined by an abrupt decreasein kidney function that includes, but is not limited to, ARF. Itis a broad clinical syndrome encompassing various etiologies,including specific kidney diseases (e.g., acute interstitialnephritis, acute glomerular and vasculitic renal diseases);non-specific conditions (e.g, ischemia, toxic injury); as wellas extrarenal pathology (e.g., prerenal azotemia, and acutepostrenal obstructive nephropathy)—see Chapters 2.2 and2.3 for further discussion. More than one of these conditionsmay coexist in the same patient and, more importantly,epidemiological evidence supports the notion that even mild,reversible AKI has important clinical consequences, includingincreased risk of death.2,5 Thus, AKI can be thought of morelike acute lung injury or acute coronary syndrome.Furthermore, because the manifestations and clinical con-sequences of AKI can be quite similar (even indistinguish-able) regardless of whether the etiology is predominantlywithin the kidney or predominantly from outside stresses onthe kidney, the syndrome of AKI encompasses both directinjury to the kidney as well as acute impairment of function.Since treatments of AKI are dependent to a large degree onthe underlying etiology, this guideline will focus on specificdiagnostic approaches. However, since general therapeuticand monitoring recommendations can be made regarding allforms of AKI, our approach will be to begin with generalmeasures.

Definition and staging of AKIAKI is common, harmful, and potentially treatable. Evena minor acute reduction in kidney function has an adverseprognosis. Early detection and treatment of AKI mayimprove outcomes. Two similar definitions based on SCrand urine output (RIFLE and AKIN) have been proposed andvalidated. There is a need for a single definition for practice,research, and public health.

2.1.1: AKI is defined as any of the following (Not Graded):K Increase in SCr by X0.3 mg/dl (X26.5 lmol/l)

within 48 hours; orK Increase in SCr to X1.5 times baseline, which

is known or presumed to have occurred withinthe prior 7 days; or

K Urine volume o0.5 ml/kg/h for 6 hours.

2.1.2: AKI is staged for severity according to the followingcriteria (Table 2). (Not Graded)

2.1.3: The cause of AKI should be determined wheneverpossible. (Not Graded)

RATIONALEConditions affecting kidney structure and function can beconsidered acute or chronic, depending on their duration.AKI is one of a number of acute kidney diseases anddisorders (AKD), and can occur with or without other acuteor chronic kidney diseases and disorders (Figure 2). WhereasCKD has a well-established conceptual model and definitionthat has been useful in clinical medicine, research, and publichealth,42–44 the definition for AKI is evolving, and theconcept of AKD is relatively new. An operational definitionof AKD for use in the diagnostic approach to alterationsin kidney function and structure is included in Chapter 2.5,with further description in Appendix B.

The conceptual model of AKI (Figure 3) is analogous tothe conceptual model of CKD, and is also applicable toAKD.42,45 Circles on the horizontal axis depict stages in thedevelopment (left to right) and recovery (right to left) ofAKI. AKI (in red) is defined as reduction in kidney function,including decreased GFR and kidney failure. The criteria forthe diagnosis of AKI and the stage of severity of AKI arebased on changes in SCr and urine output as depicted in thetriangle above the circles. Kidney failure is a stage of AKIhighlighted here because of its clinical importance. Kidneyfailure is defined as a GFR o15 ml/min per 1.73 m2 body

http://www.kidney-international.org c h a p t e r 2 . 1

& 2012 KDIGO

Table 2 | Staging of AKI

Stage Serum creatinine Urine output

1 1.5–1.9 times baselineOR

X0.3 mg/dl (X26.5mmol/l) increase

o0.5 ml/kg/h for6–12 hours

2 2.0–2.9 times baseline o0.5 ml/kg/h forX12 hours

3 3.0 times baselineOR

Increase in serum creatinine toX4.0 mg/dl (X353.6mmol/l)

ORInitiation of renal replacement therapyOR, In patients o18 years, decrease ineGFR to o35 ml/min per 1.73 m2

o0.3 ml/kg/h forX24 hours

ORAnuria for X12 hours

Kidney International Supplements (2012) 2, 19–36 19KDIGO,KI,2012

Nature Reviews | Nephrology

Stage 0 subtypesC: SCr not back to

baseline

B: Biomarker or loss of

renal reserve

indicates injury

A: No evidence of injury

Injury

Up to 7 days 7–90 days >90 days

AKI KDIGO stage

Ongoing RRT

3 (SCr 3x)/RRT

2 (SCr 2x)

1 (SCr 1.5x)

Subacute AKI

AKD stage (congruent to AKI stage) CKD

Ongoing RRT

3 (SCr 3x)/RRT

2 (SCr 2x)

1 (SCr 1.5x)

0 Subacute AKD

broader population of patients with AKD. Further work is warranted to delineate the epidemiology of AKD, including differences in the predictors, course, and outcomes relative to AKI. Few data exist on char-acterizing the phases of AKD, including the processes by which patients recover or progress to CKD, the evolving risk experienced by AKD survivors, and the processes of care experienced.

For the purposes of our recommendations, AKD is conceptualized not as pre-CKD but rather, as post-AKI. This distinction has important implications for the diag-nosis, care and follow-up of affected patients, including the notion that AKD might exist even in the absence of standard clinical evidence (FIGS 2,4).

The ideal definition for recovery should quantify lost pre-existing kidney function as well as current residual kidney function and reserve, identify when recovery is complete, and provide prognostic information (BOX 3). Intrinsic to the concept of AKD is that acute loss of kid-ney function or damage extends beyond diagnosis and staging of AKI and highlights additional points of poten-tial intervention from the onset of injury through to the more convalescent phase of disease that could modify long-term outcomes. No standardized definition of recovery from AKI or AKD exists, and only a few studies have evaluated the kinetics or trajectory of recovery from either AKI or AKD among patients not on dialysis (see Supplementary information S3 (table)). Although these studies have used varying time frames and thresholds of serum creatinine level to define recovery, the results generally show a graded association between recovery and future risk of mortality, loss of kidney function, and other morbidities.

Other potential measures of recoveryAKD and recovery from AKD are currently assessed using filtration markers, such as serum creatinine. This approach has limitations, however, and loss of mus-cle mass, changes in volume of distribution, changes

in renal reserve, and hyperfiltration can confound the assessment of functional recovery54–60. The limitations of using serum creatinine to assess recovery are supported by observational data indicating that AKI is associated with an increased risk of CKD, even when accompanied by an apparent complete return of serum creatinine to baseline levels61,62.

Alternative or complementary measures of kidney function, including other filtration markers such as cystatin C and timed urine clearance measurements, could hold promise for improved phenotyping of func-tional recovery from AKD but require further validation before recommending their routine adoption into clinical practice35,63–66. Methods to assess glomerular functional reserve (for example, by assessing the effect of a protein load on GFR) or tubular functional reserve (for example, through furosemide stress testing or the administration of intravenous creatinine) have also been developed in the CKD setting but have yet to be applied to patients with AKD67,68. Interestingly, serum creatinine level has been the standard approach to the assessment of renal function for decades, but intravenously administered creatinine fails to increase GFR in humans, regardless of renal function68. Intravenous creatinine does, however, significantly increase creatinine clearance68, demonstrat-ing that glomerular and tubular reserve do not necessar-ily correlate and suggesting that patients with CKD can maintain some preservation of glomerular renal reserve but fail to show any measurable tubular reserve68–71. On the basis of these findings, assessments of glomerular and tubular reserve are likely to assess different facets of kid-ney disease. Several studies have also examined the use of next-generation biomarkers of tubular injury and furo-semide stress testing to predict recovery from AKI72,73. As many of these markers reflect ongoing tubular injury, most studies have focused on their ability to indicate the likelihood of recovery during early or peak AKI in select groups of patients (see Supplementary information S4 (table)). Further work is needed to determine the utility of these biomarkers in informing clinical decision-making.

A framework to classify AKD and recoveryA useful classification of recovery from AKD would quantify the extent to which kidney function was lost, indicate when repair is complete and damage is no longer occurring, provide a measure of a patient’s current kidney function and reserve, and provide prognostic information. A scheme that aligns with and integrates the KDIGO cate gories for AKI and provides a simple and translatable framework for ascertaining transition points for outcomes during AKD and at the end of 90 days would be ideal. Accordingly, we propose to map the KDIGO AKI stag-ing categories to the staging of AKD for the purpose of defining the severity of AKD and to offer a framework for kidney-specific outcomes across a 90-day timeline (FIG. 4). In this conceptual framework, improvements in kidney function and/or resolution in kidney damage would be staged by an improvement (decrease) in AKD stage (for example, a shift from stage 3 AKD to stage 2 or lower). We recognize that specific thresholds to define ‘recovery’ remain to be defined, in particular in selected populations

Figure 4 | Interplay between acute kidney injury (AKI), acute kidney disease (AKD) and chronic kidney disease (CKD). AKI stages map directly to the new proposed AKD stages. In addition, patients with AKD can progress to CKD. Stage 0 AKD represents partial recovery from AKI. Stage 0C includes patients for whom serum creatinine levels are higher than baseline but within 1.5 times baseline levels. Stage 0B includes patients whose serum creatinine has returned to baseline levels, but who still have evidence of ongoing kidney damage, injury, or loss of renal reserve. Stage 0A includes those patients who have had an episode of AKI and retain a risk of long-term events without structural or damage markers for AKD. Patients whose serum creatinine level has not returned to baseline and who have ongoing evidence of kidney damage and/or injury are termed stage 0B/C.

C O N S E N S U S S TAT E M E N T

NATURE REVIEWS | NEPHROLOGY ADVANCE ONLINE PUBLICATION | 7

such as survivors of critical illness or among patients who no longer fulfil criteria for AKI or AKD stage 1 but whose serum creatinine level has not yet returned to baseline62,74. In this framework, we propose a ‘stage 0ʹ, with A, B, and C subgroups (TABLE 1). Stage 0C includes patients for whom serum creatinine levels are higher than baseline but within 1.5 times baseline levels. Population studies suggest that these patients who achieve a recovery serum creatinine level that remains above 115% of baseline levels still carry a mortality risk74. Thus, these patients with AKD might require further follow-up and could be candidates for future therapeutic intervention. Stage 0B includes patients whose serum creatinine has returned to their baseline level after an episode of AKI, but still have evidence of ongoing kidney damage. The diagnosis of this ongoing damage for most patients will likely be in the form of new-onset proteinuria, worsened proteinuria from base-line, new-onset hypertension, or worsening hypertension. In addition to proteinuria and hypertension, evidence of ongoing kidney disease might be assessed through use of biomarkers or imaging studies. Stage 0B also includes patients for whom serum creatinine level has returned to baseline after an episode of AKI with no evidence of ongoing kidney damage, but who have experienced a loss of renal reserve. One example of this scenario would be a patient who has undergone a nephrectomy, whereby the contralateral kidney might adapt to the loss of renal mass, but a significant portion of renal reserve has nonetheless been lost. The assessment of renal reserve can be assessed by both glomerular and tubular stress testing75. Patients in whom serum creatinine levels fail to return to baseline and have evidence of ongoing injury would be classified as having Stage 0B/C. As the study of AKD is nascent, future research should carefully assess the risk of future events associated with these AKD stages (BOX 4). In addi-tion, the thresholds of the various biomarkers, imaging outputs, and/or renal reserve that define ‘full recovery’ versus ongoing risk is not known and will require further investigation. Stage 0A encompasses patients who have no evidence of damage or functional loss following an AKI episode and represents clinical recovery. These patients may nonetheless be vulnerable to further kidney damage and other adverse events. As has been shown previously,

patients who have suffered an AKI event and ‘recover’ still carry a long-term increased risk of major adverse cardiac and kidney events49,76. Patients with AKD stage 0A might still require follow-up and could likely benefit from avoid-ing unnecessary nephrotoxic drugs. We hypothesize that this framework will enable the recognition and descrip-tion of the dynamic nature of AKI and AKD beyond the initial diagnosis and staging of kidney injury, which will enable improved understanding of the natural course of the disease and ultimately facilitate the development of specific care pathways to guide surveillance, investi-gation and interventions, and align with care beyond 90 days. However, the accuracy and usefulness of these proposed stages in assessing kidney function and damage in patients with AKD requires further validation.

Finally, in keeping with the original conceptual framework for AKD as proposed by KDIGO, we rec-ognize that AKI might not have always been diagnosed in a patient who appears to have an acute deterioration in renal function. In other words, the diagnosis of AKD may require inference of the existence of an episode of AKI. For example, consider a patient who is seen for an annual internist visit. The patient’s serum creatinine is found to be twice the level observed the year before and they describe a severe ‘flu-like’ illness 2 months prior that lasted a week but eventually resolved without medical attention. We would suggest that treating this situation as a likely case of AKD — for example, by requesting that the patient avoids unnecessary nephrotoxins, requesting follow-up serum creatinine measurements, and screening for CKD risk factors — would be reasonable.

Follow-up careAs the evidence linking AKI with loss of kidney func-tion45–47,77–80, hypertension81,82, cardiovascular dis-ease49,50,83, and death46,83–87 accumulates, determining the optimal care for this growing population is critical. The American Society of Nephrology AKI Advisory Group has highlighted the transition of care as a poten-tial opportunity to reduce the long-term impact of AKI88, and hence, AKD. However, a paucity of data exists to indicate which interventions can reduce morbidity and mortality in AKI/AKD survivors.

Table 1 | Recommendations for AKD staging

Stage Definition

Stage 0* A: Absence of criteria for B or C.

B: Continued evidence of ongoing injury, repair and/or regeneration or indicators of loss of renal glomerular or tubular reserve

C: Serum creatinine level

Chawla,NatRev,2017

were 64, 78, 50, and 18%, respectively. The survival curves ofthese four groups of patients had statistically significantdifferences (log-rank P¼ 0.000) (Figure 2a). There were nostatistical differences between the survival curves of the groupof patients younger than 31 years and that of patients withages ranging between 31 and 50 years (P¼ 0.172). The

comparisons between the survival curves of the other agegroups are shown in Figure 2a. A similar analysis groupingthe patients by decades gave similar results. The survivalcurve of each decade was worse than that of the precedingone with the exception of the patients younger than 21 whohad a worse survival curve than that of the patients with agesranging between 21 and 30 years of age (not shown).

The probability of survival at 10 years after discharge forpatients having a trauma, surgical or medical admission was84, 53, and 44%, respectively. The survival curves of thesethree different types of admission were statistically different(log-rank P¼ 0.041) (Figure 2b).

The survival curve of patients without comorbid factorswas significantly better (log-rank P¼ 0.000) than that ofpatients with associated comorbid processes (Figure 2c). At10 years after discharge, the survival rate was 85% in the firstgroup and 42% in the second.

When we analyzed individually the survival curves of thedifferent comorbid pathologies considered in this study, nostatistical difference was observed among them (data notshown). This could be due to the small number of patients ineach subgroup. Also, the fact that 48% of the patients had atleast two comorbid factors could have contributed to thesefigures.

During the follow-up 95 patients died. Ninety-one ofthem (97%) had comorbidity factors at the time of ARF. Thecauses of death among the four patients without comorbidityfactors were an acute myocardial infarction, urinary sepsisassociated with a permanent urinary catheter, drug overdose,and one unknown. At the time of death, two of these patientshad normal renal function and the others had mild renalinsufficiency.

The survival curve of patients treated in an intensive careunit (ICU) setting was significantly better (log-rankP¼ 0.002) than that of patients treated in other hospitalareas. At 10 years the survival rate was 61 and 42% for ICUand non-ICU patients, respectively (Figure 2d).

Long-term outcome, assessed by the survival curves, wasbetter among those patients who at the time of hospitaldischarge had completely recovered their renal function thanamong those with only partial recovery (log-rank P¼ 0.028)

Table 1 | Characteristics of the survivor patients at themoment of the acute tubular necrosis renal failure episode

Patients

(No=187) (%)

Age/years 57.8716.2Gender (female/male) 63/124 34/66Comorbidity factors (no/yes) 35/152 19/81

Type of admissionMedical 95 51Surgical 81 43Trauma 11 6

ATN ethiologyNephrotoxic 97 52Sepsis 26 14Medical 32 17Surgical 32 17

Severity Index (SI) 0.3170.2

ARF functional severityMild 20 11Moderate 79 42Severe 88 47

Type of ARFOliguric 72 39Non oliguric 115 61

DialysisYes 57 30No 130 70

ICU admissionYes 80 43No 107 57

ICU stay (days) 14.7714.0

Recovery of renal function with respect to serum creatinine at admissionTotal recovery 78 42Partial recovery 107 57No recovery 2 1

Renal function at dischargeNormal renal function 78 42Mild renal insufficiency 64 13Moderate renal insufficiency 39 21Severe renal insufficiency 6 3Dialysis 0

Final serum creatinine (mg/dl) 1.770.7Length of admission (days) 53.9736.9Length of nephrological care (days) 18.9713.6

ARF, acute renal failure episode; ATN, acute tubular necrosis; ICU, intensive care unit;SI, severity index.Continuous variables are expressed as mean7s.d.

0.9

1.0

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

01

187 153 140 128 123 117 110 97 87 73 63 50 45 34 29 25 21 14 12 8 5 3 1

2 3 4 5 6 7 8 9 10 11YearsPatients

(N)

12 13 14 15 16 17 18 19 20 21 22

Cum

ulat

ive

surv

ival

rate

Figure 1 |Kaplan–Meier survival curve of the patients dischargedalive after an ATN episode.

680 Kidney International (2007) 71, 679–686

or ig ina l a r t i c l e F Liaño et al.: Long-term outcome of acute tubular necrosis

were 64, 78, 50, and 18%, respectively. The survival curves ofthese four groups of patients had statistically significantdifferences (log-rank P¼ 0.000) (Figure 2a). There were nostatistical differences between the survival curves of the groupof patients younger than 31 years and that of patients withages ranging between 31 and 50 years (P¼ 0.172). The

comparisons between the survival curves of the other agegroups are shown in Figure 2a. A similar analysis groupingthe patients by decades gave similar results. The survivalcurve of each decade was worse than that of the precedingone with the exception of the patients younger than 21 whohad a worse survival curve than that of the patients with agesranging between 21 and 30 years of age (not shown).

The probability of survival at 10 years after discharge forpatients having a trauma, surgical or medical admission was84, 53, and 44%, respectively. The survival curves of thesethree different types of admission were statistically different(log-rank P¼ 0.041) (Figure 2b).

The survival curve of patients without comorbid factorswas significantly better (log-rank P¼ 0.000) than that ofpatients with associated comorbid processes (Figure 2c). At10 years after discharge, the survival rate was 85% in the firstgroup and 42% in the second.

When we analyzed individually the survival curves of thedifferent comorbid pathologies considered in this study, nostatistical difference was observed among them (data notshown). This could be due to the small number of patients ineach subgroup. Also, the fact that 48% of the patients had atleast two comorbid factors could have contributed to thesefigures.

During the follow-up 95 patients died. Ninety-one ofthem (97%) had comorbidity factors at the time of ARF. Thecauses of death among the four patients without comorbidityfactors were an acute myocardial infarction, urinary sepsisassociated with a permanent urinary catheter, drug overdose,and one unknown. At the time of death, two of these patientshad normal renal function and the others had mild renalinsufficiency.

The survival curve of patients treated in an intensive careunit (ICU) setting was significantly better (log-rankP¼ 0.002) than that of patients treated in other hospitalareas. At 10 years the survival rate was 61 and 42% for ICUand non-ICU patients, respectively (Figure 2d).

Long-term outcome, assessed by the survival curves, wasbetter among those patients who at the time of hospitaldischarge had completely recovered their renal function thanamong those with only partial recovery (log-rank P¼ 0.028)

Table 1 | Characteristics of the survivor patients at themoment of the acute tubular necrosis renal failure episode

Patients

(No=187) (%)

Age/years 57.8716.2Gender (female/male) 63/124 34/66Comorbidity factors (no/yes) 35/152 19/81

Type of admissionMedical 95 51Surgical 81 43Trauma 11 6

ATN ethiologyNephrotoxic 97 52Sepsis 26 14Medical 32 17Surgical 32 17

Severity Index (SI) 0.3170.2

ARF functional severityMild 20 11Moderate 79 42Severe 88 47

Type of ARFOliguric 72 39Non oliguric 115 61

DialysisYes 57 30No 130 70

ICU admissionYes 80 43No 107 57

ICU stay (days) 14.7714.0

Recovery of renal function with respect to serum creatinine at admissionTotal recovery 78 42Partial recovery 107 57No recovery 2 1

Renal function at dischargeNormal renal function 78 42Mild renal insufficiency 64 13Moderate renal insufficiency 39 21Severe renal insufficiency 6 3Dialysis 0

Final serum creatinine (mg/dl) 1.770.7Length of admission (days) 53.9736.9Length of nephrological care (days) 18.9713.6

ARF, acute renal failure episode; ATN, acute tubular necrosis; ICU, intensive care unit;SI, severity index.Continuous variables are expressed as mean7s.d.

0.9

1.0

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

01

187 153 140 128 123 117 110 97 87 73 63 50 45 34 29 25 21 14 12 8 5 3 1

2 3 4 5 6 7 8 9 10 11YearsPatients

(N)

12 13 14 15 16 17 18 19 20 21 22

Cum

ulat

ive

surv

ival

rate

Figure 1 |Kaplan–Meier survival curve of the patients dischargedalive after an ATN episode.

680 Kidney International (2007) 71, 679–686

or ig ina l a r t i c l e F Liaño et al.: Long-term outcome of acute tubular necrosis

insufficiency), two patients had moderate renal failure (athospital discharge one was in the same situation and theother had mild renal insufficiency) and finally one patientwas in periodic hemodialysis (at hospital discharge he hadmoderate renal failure). With the exception of the patient onchronic hemodialysis, the analytical data of these patients areshown in Table 3.

All the inhabitants X15 years of age living in Madrid on 1January 1978 were studied as a cohort. The survival curve ofthis cohort is shown in Figure 3a and compared graphicallywith that of the patients who survived an ATN episode. It canbe clearly appreciated that the slope of the curve of patientshaving had ARF is worse than that observed in the generalpopulation. A similar analysis, sub-classifying the patientsand the general population in three subgroups of age, alsodemonstrated that the poorer vital expectative of patientswith ARF in comparison with the general population was notconditioned by an age effect (Figure 3b).

DISCUSSIONMethodologically, as the ARF syndrome has many etiologies,we limited the study only to patients having ATN owing tothe fact that: (1) ATN is the paradigmatic form of thesyndrome;14 (2) it is the most frequent type of ARF observedin developed countries,5,15 and (3) it would allow us tomanage a more homogeneous population for analysis. Toavoid methodological bias, in the analysis we only includedpatients with a SCro1.4mg/dl and ARF in organ transplantpatients was excluded. The methods used were singular withrespect to: (1) the period of inclusion, 15 years; (2) theprolonged period of potential follow-up of the patients, from7 to 22 years; (3) the number of patients analyzed; (4) thesmall percentage of patients with unknown evolution; and(5) the fact of studying not only the survival outcome butalso the renal functional outcome. In fact, on analyzing theliterature (Table 4) we found that: (1) the long-term follow-up period of the patients studied is very variable, ranging

Table 3 | Analytical data of the acute tubular necrosis surviving patients in their last clinical evaluation during 2000–2001

Parametera Normal renal function (n=46) Mild renal insufficiency (n=9) Moderate renal insufficiency (n=2)

Plasma (n=57)Creatinine (mg/dl) 1.17 0.2 1.770.1 3.371.5Urea (mg/dl)11 45.3735.0 75.1735.7 103.0721.2Na (mEq/l) 140.072.3 138.876.1 142.372.8K (mEq/l) 4.470.5 4.470.3 4.670.5Cl (mEq/l) 105.774.0 102.573.7 105.570.7Total CO2 (mEq/l) 28.273.0 28.172.0 24.673.6Uric acid (mg/dl) 5.971.4 6.772.2 9.273.5Glucose (mg/dl) 115.2746.3 113.4726.2 118.2733.5Total proteins (mg/dl) 7.370.7 7.171.3 7.670.3Hemoglobin (g/dl) 1571.4 13.973.2 11.671.3Hematocrit (%) 44.374.0 41.873.0 36.175.8

Urine (n=40)Creatinine (mg/dl) 104,7749.0 71.0729.0 45.5714.8Urea (g/l) 18.676.0 11.275.0 7.172.0Na (mEq/l) 116.071.0 88.774.3 76.974.6K (mEq/l) 39.071.9 30.072.2 18.072.6Cl (mEq/l) 100.5740.0 94.6741.0 67.8735.0Proteinuria* (mg/min) 0,4371.18 (n=12)** 0.4370.45 (n=4) 0.5470.03 (n=2)Creatinine clearance (ml/min) 101.3726.4 (n=29) 52.1712.1 (n=9) 37.372.5 (n=2)

To convert urea to BUN values, in mg/dl, divide by 2.14.aValues are expressed as mean7s.d.*Means estimate only the values of patients with positive proteinuria.**This group includes several patients with diabetes, one of them with nephrotic proteinuria.

Table 2 | Should be rebuilt as follows

Reference value Regression coefficient P HR CI

Age 0.04 0 1.038 1.02–1.06

Type of admission: 0Surgical Medical admission !0.8 0 0.45 0.29–0.70Trauma !0.88 0.397 0.42 0.05–3.16

Presence of Comorbidity factors Absence of comorbidity 1.46 0.006 4.29 1.51–12.17Sex Female 0.54 0.02 1.72 1.09–2.72

CI, confidence interval; HR, hazard ratio.Log likelihood ratio: X2=52.59; P=0.000.

682 Kidney International (2007) 71, 679–686

or ig ina l a r t i c l e F Liaño et al.: Long-term outcome of acute tubular necrosis

Liano,KI,2007

Totalrecovery: SCreat <123umol/lPartialrecovery: 123<SCreat <maxScreatAbsenceofrecovery: SCreatmax

reported by Frost et al.2 Additionally, Gentric and Cledes15 ina study of renal function after ATN in patients o 65 years,reported that none of their patients died owing to renalinsufficiency during the long-term outcome. Although, fromour data, we could not conclude whether prevention of ATNin patients with comorbid conditions could play a favorablerole in the long-term outcome of this condition, it seemsreasonable to implement prophylactic measures in patients insituations of risk for developing ARF.

As in other series,11,12 renal function measured by SCrimproved during the first 6 months after discharge. Afterthat, SCr tended to stabilize, although with wide dispersionin individual values. In agreement with these data, the lastclinical evaluation in 57 of our patients was satisfactory inmost cases although, at some stage during the follow-up,1.6% of the patients required dialysis treatment.

The graphic comparison of the survival curves of the ARFpatients with the cohort of the whole adult population of the1978 Madrid in-habitants discloses the importance of thelong-term outcome of ARF by itself. At 10 years, the survivalrate of the Madrid population was 92%, whereas that of theARF patients was 50%. Frost et al.2 compared the survival oftheir patients with medical or surgical ARF and that of aDanish-matched population. However their data are notdirectly comparable with our results as they studied theoutcome from the moment of hospital admission, whereaswe only considered long-term outcome of patients dis-charged alive after an ATN episode. The poorer survival ofthe ARF patients in comparison with the Madrid populationcannot be attributed to age as is clearly shown in Figure 3. Tothe best of our knowledge this is the first time that this typeof analysis has been reported.

Table 4 | Literature review: long-term outcomes of patients surviving an ARF

Number ofpatients(characteristics)

Periodstudied ARF etiology

Survivors (n)in-hospital

mortality (%)

Maximumfollow-upin months

Number oflong-termstudiedpatients

Follow-uprate

Long-termsurvival (%)

Functionaloutcome

Turney et al.1a 142 Severe ARF 1956–1987/R Obstetriccauses

112 (21%) 372 102 91% 1 year. 79%h

5 years: 75%h

10 years: 72%h

SCrs increased:48%

Frost et al.2b 419 (100% dialysis) 1977–1988/R 82% ATN 226 (46%) 144 226d (73%ATN)

100% Medical ARF:52% (5 years)g

Surgical ARF: 28%(5 years)f

NR

Chertow et al.3b 132 ICU setting(100% dialysis)

1991–1993/R 78% ATN 40 (70%) 12 39 97% 81%f Chronic dialysis:33%h

Lameire et al.4b 230 ICU setting(100% dialysis)

1993–1995/R Most ATN 79 (66%) 18 79e NA 15%g NR

Brivet et al.5c 360 ICU setting(48% dialysis)

1991/P 78% ATN 150 (58%) 6 86 57% 87% (6 months)h SCrX1.7mg/dl:28%

McCarthy et al.6c 142 ICU setting(71 patients ineach period)

1977–1979and

1991–1992/R

480% ATN 23 (68%) 12 23 100% Period 77–79:21%g

Period 77-79:Chronic dialysis:4%

480% ATN 37 (48%) 12 37 Period 91-92:30%g

Period 91–92:Chronic dialysis:21%

Khan et al.7b 310(Epidemiological –community-based)

1989–1990/R Miscellaneous NR(?) 24 NR NA 31% (2 years)g NR

Korkeila et al.8b 62 ICU setting(100% dialysis)

1992–1993/R ATN 34 (45%) 60 3411 100% 35% (5 years)g Chronic dialysis:8%g

Stevens et al.9c 288(Epidemiological –community-based)

1996/P Miscellaneous 161 (56%) 36 161 100% 28% (3 years)g NR

Morgera et al.10b 979 (100% CRRT) 1993–1998/R Probablymainly ATNf

301 (69%) 88 267 89% 50% (5 years)h Chronic dialysis:10%h

Ramón y CajalHospitalexperiencea

413 General series(48% dialysis)

1977–1992/R ATN 187 (55%) 264 177 95% 50% (10 years)h Chronic dialysis:2%h

ATN, acute tubular necrosis; CRRT, continuous renal replacement therapy; ICU, intensive care unit; NA, not assessable; NR, not reported; P, prospective; R, retrospective;SCr, serum creatinine.Follow-up rate: number of long-term followed-up patients with respect to the number of patients discharged alive.aCases with previous renal function impairment were excluded.bPrior renal function unknown.cIncludes patients with prior renal insufficiency.dStill alive according to the Danish Central Personal Register.eAccording to the Finnish National Population Register.fPersonal interpretation of the authors’ data.gWith respect to the initial population.hWith respect to the patients discharged alive.

684 Kidney International (2007) 71, 679–686

or ig ina l a r t i c l e F Liaño et al.: Long-term outcome of acute tubular necrosis

Liano,KI,2007

1088

Data of renal function during routine follow-up, in the outpatientclinic, of 14 patients who had survived severe episodes of acute tubularnecrosis. A few of the clearances in the acute phase of the illnessare also included to illustrate the pattern of recovery of renal function.

after her episode of renal failure. 3 patients had mildattacks of polyarthritis in their first year of follow-up,and 1 of these had a severe attack of " atypical pneu-monia " later. Minor symptoms were mild headachesand, in 1 patient, nocturia. There were no noteworthyfindings on physical examination except that some of therandom blood-pressures were above the normal range.The significance of this was difficult to assess, because thecircumstances of outpatient attendance were not such asto provide basal conditions.

Renal-function tests were made as convenient. Tech-nical methods were as described elsewhere (Bull et al.1950), except that the urine was not collected with acatheter. Since this may reduce the accuracy of thesetests, they are presented’ not as exact indices of renalhaemodynamics but as simple tests of renal function. Theaccompanying figure and the table show that thereusually is good recovery of renal function within the firstsix months, and that this is sustained subsequently.Renal function, however, tends to remain below thelower limit of normal. In no case was there persistentalbuminuria or pathological deposit microscopically.

DISCUSSION

This follow-up report seems to indicate that, oncerecovery has been made from the acute episode ofacute tubular necrosis, a favourable prognosis can begiven. A lengthy convalescence of from three to sixmonths may be advisable, because this is the time takenfor renal function to approach the lower limit of normal,although physical well-being returns much more quickly.The slightly subnormal renal function in the late follow-upperiod is of interest. It may well be due to residua of theacute episode (scarring, vascular damage, &c.), but thisremains open to speculation. At any rate, the levels ofrenal function attained are compatible with normalexpectation of life, although the renal reserve isdiminished. The occasional random blood-pressurereadings outside normal limits invite speculation aboutwhether hypertension will occur with undue frequencyamong such patients, but a follow-up of more patientsand for a much longer time is required to answer thisquestion.

This unit’s policy was to allow subsequent pregnanciesin these patients. As already stated, 1 patient wasobserved during and after a normal pregnancy, and, ofthe other patients who recovered but were unable to

attend as outpatients, 2 had subsequent normalpregnancies.

SUMMARY

There has previously been doubt about the degree ofrecovery from acute tubular necrosis, and no previouslong-term follow-up has been reported..An interim follow-up report on 14 patients, some of

whom have been seen for three years, is presented.Good clinical recovery, which is sustained, is the rule.The question of increased frequency of hypertension

remains open.Renal function becomes adequate but tends to remain

below the lower limit of normal.Pregnancy seems to be well tolerated subsequently.Further follow-up of more patients and for a much

longer time is required.I am indebted to Dr. G. M. Bull for the opportunity to

study patients under his care, and to Miss Barbara Evans,B.se., for technical assistance.

REFERENCES

Bull, G. M., Joekes, A. M., Lowe, K. G. (1950) Clin. Sci. 9, 379.Burch, G. E., Ray, C. T. (1949) Ann. intern. Med. 31, 750.Burwell, E. L., Kinney, T. D., Finch, C. A. (1947) New Engl. J. Med.

237, 657. Lucké, B. (1946) Milit. Surg. 99, 371. Marshall, D., Hoffman, W. S. (1949) J. Lab. clin. Med. 34, 31.Sirota, J. H. (1949) J. clin. Invest. 28, 1412.

WEIL’S DISEASEA CAUTIONARY TALE

P. J. O’CONNELLM.D. Durh., B.Hy., D.P.H.

MEDICAL OFFICER OF HEALTH, BOROUGH OF BEDDINGTONAND WALLINGTON

J. C. BROOMM.D. St. And.

BACTERIOLOGIST, WELLCOME RESEARCH INSTITUTION, LONDON

THIS investigation was undertaken as the result ofa rash opinion expressed by one of us (J. C. B.), andbecause, at first, the mistaken diagnosis seemed to beconfirmed by our inquiries. The error became apparentwhen more information was obtained, but we thinkthat our findings have an intrinsic interest in additionto the warning they give against " spot diagnosis."

INTRODUCTION

In the early stages of leptospirosis a patient’s serumnot infrequently reacts with more than one species ofleptospire ; indeed Gispen and Schunner (1939) foundthat the titre against the infecting species might be lowerthan the titre for some related serological type. Of lesscommon occurrence is what Fuhner (1951) calls the" paradoxical reaction," in which antibodies for a,heterologous species appear, sometimes in high con-centration,’ before any positive reaction can be demon-strated with the homologous leptospire. As the diseaseprogresses the antibody level for the causal organismusually rises, while the titres for the heterologous speciesfall. But heterologous titres may remain at a highlevel until well into convalescence or even after recovery,and in such cases the diagnosis can be established onlywith the help of cross-absorption tests. ’.At the beginning of an illness, therefore, it may be

possible to state only that the patient is suffering fromleptospirosis. Thus, any device is useful if it leads toan early, complete diagnosis, because identification ofthe species is not merely of academic interest. If, forexample, the causal organism is Leptospira canicola, thedog which was the source of infection may transmit thedisease to other humans unless it is traced and renderedinnocuous.

SEROLOGICAL INVESTIGATION

A boy, aged 15, fell ill on Aug. 8, 1951, and wasthought by his doctor to be suffering from meningitis.

1086

after an injection of sodium salicylate 100 mg. per kg.body-weight, the amount of ascorbic acid and cholesterolin the adrenals decreased and nuclear pyknosis appeared,but there was no significant change in the number ofcirculating eosinophils. After an injection of 50 mg. per kg.,the amount of ascorbic acid did not decrease, but that ofcholesterol did, and pyknosis appeared. After an injectionof 25 mg. per kg., pyknosis was the only evidence ofactivity of the pituitary-adrenal axis. It must beemphasised that nuclear pyknosis is produced by anadrenal action, for it does not occur in adrenalectomisedanimals, even when a large dose of salicylate is injected.

Thus, independently of the changes which might occurthrough the handling of the rats, the intraperitonealinjection of a small dose of sodium salicylate stimulatedthe adrenals-the evidence for this being nuclear pyk-nosis in

lymphoidtissue.Kelemen etal. (1950),studyingthe effect ofsodium sali-cylate onthe circula-ting eosino-phils, didnot observea significantdecrease inthe numberof circula-ting eosino-phils whenthey injec-ted intorats dosessmallerthan 300

mg. per kg. body-weight ; but these workers have notdescribed the conditions of their experiments.

Hetzel and Hine (1951) found that even an injectionof sodium salicylate 50 mg. per kg. body-weight depletedthe amount of ascorbic acid in the adrenals, but theyused a hypertonic solution (personal communication)and the rats were injected on several days.Our observations seem much more conclusive. They

helped us to understand the dissociation of the bio-chemical, hematological, and histological signs of activityof the pituitary-adrenal axis under dial anaesthesia.No difference in response was observed between dial-

ansesthetised rats given sodium salicylate 500 mg. perkg. body-weight and normal rats given 50 mg. per kg.Dial therefore seems to inhibit the adrenal response tosalicylates. As barbiturates exert their effects principallyon the hypothalamus (Fulton 1949), this inhibitionsuggests that the action of the salicylates is exerted onthe hypothalamus.,

, CONCLUSIONS

(1) Signs of adrenal activity are produced by a doseof sodium salicylate smaller than that which producesthe high blood-salicylate level regarded by Coburn(1943) as necessary for the treatment of acute rheumatism.Low blood-salicylate levels (6 or 7 mg. per 100 ml.)cause nuclear pyknosis in the lymphoid tissue of rats.

(2) Injection of sodium salicylate reveals a gradientamong the responses generally regarded as evidence ofadrenal stimulation.

(3) The response to injection of sodium salicylate issmaller in rats anaesthetised with dial.

(4) Inhibition of the pituitary-adrenal mechanism bybarbiturates may be important in the treatment ofcertain endocrine disorders.

Fig. 8-Changes in blood-salicylate level, amounts ofcholesterol and ascorbic acid in adrenals, and number

ofcirculatingeosinophils in relation to dose ofsodiumsalicylate.

Fig. 9--Section of lymph-node showing nuclear pyknoses 4 hours afterinjection with sodium salicylate 25 mg. per kg. body-weight in non-anaesthetised rat.

(5) In the treatment of rheumatic disease it seemsinadvisable to administer salicylates and large doses ofbarbiturates together.We are indebted to the Belgian Foundation for Scientific

Research for grants in aid of this investigation.

REFERENCES

Betz, H., Van Cauwenberge, H. (1951) C.R. Soc. Biol. Paris, 145,1275, 1439.

Champy, C., Demay, M. (1951) Bull. Acad. Méd. Paris, 135, 13.Coburn, A. F. (1943) Bull. Johns Hopk. Hosp. 73, 435.Fulton, J. F. (1949) Textbook of Physiology. 16th ed., Philadelphia

and London ; p. 239.Harris, G. W. (1950) Abstr. Int. physiol. Congr. p. 247.Hetzel, B. S., Hine, D. (1951) Lancet, ii, 94.Hume, D. W. J. (1949) J. clin. Invest. 28, 790.Kelemen, E., Majoros, M., Ivanyi, J., Kovacs, K. (1950) Erperientia,

6, 435.Robinson, F. B. (1951) Brit. med. J. i, 300.Roskam, J., Van Cauwenberge, H., Mutsers, A. (1951) Lancet, ii, 375.Van Cauwenberge, H. (1951) Ibid, 374, 686.

— Heusghem, C. (1951a) Ibid, i, 771.— — (1951b) C.R. Soc. Biol. Paris, 145, 1272.— — (1952) Acta med. scand. 141, 265.

THE LATE PROGNOSIS IN ACUTETUBULAR NECROSIS

AN INTERIM FOLLOW-UP REPORT ON 14 PATIENTS

K. G. LOWEM.D. St. And., M.R.C.P.

LATE MEDICAL REGISTRAR AND TUTOR IN MEDICINE,POSTGRADUATE MEDICAL SCHOOL OF LONDON

"ACUTE tubular necrosis " is synonymous withLucke’s (1946) -term of " lower-nephron nephrosis "and includes . a wide group of anurias following suchvaried causes as ingested poisons, mismatched blood-transfusion,. abortion, and concealed accidental hæmor-rhage. This group of anurias has probably always beenof common occurrence, but it is only in the last decadeor so that it has been adequately described and clearly.recognised as a -clinical and pathological entity. Theextensive literature that has accumulated on the subjectdeals only with the acute episode of renal failure. AsBurch and Ray (1949) point out, no adequate long-termfollow-up of a series of cases has yet been reported.

. Marshall and Hoffman (1949) showed that in’3 patientscomplete functional restitution of the kidneys to normaloccurred, taking three months in 1 patient and nearlyseven months in the other 2 patients. In a. 4th patient

Lowe,Lancet,1952

AKIETMRC INCIDENTE

Acute Kidney Injury Increases Risk of ESRD amongElderly

Areef Ishani,* Jay L. Xue,* Jonathan Himmelfarb,† Paul W. Eggers,‡ Paul L. Kimmel,‡§

Bruce A. Molitoris,! and Allan J. Collins*

*United States Renal Data System Coordinating Center and Department of Medicine, University of MinnesotaMedical School, Minneapolis, Minnesota; †Division of Nephrology, Maine Medical Center, Portland, Maine;‡National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda,Maryland; §Division of Renal Diseases and Hypertension, Department of Medicine, George Washington UniversityMedical Center, Washington, DC; and !Division of Nephrology, Indiana University School of Medicine, Indianapolis,Indiana

ABSTRACTRisk for ESRD among elderly patients with acute kidney injury (AKI) has not been studied in a large,representative sample. This study aimed to determine incidence rates and hazard ratios for developingESRD in elderly individuals, with and without chronic kidney disease (CKD), who had AKI. In the 2000 5%random sample of Medicare beneficiaries, clinical conditions were identified using Medicare claims;ESRD treatment information was obtained from ESRD registration during 2 yr of follow-up. Our cohortof 233,803 patients were hospitalized in 2000, were aged !67 yr on discharge, did not have previousESRD or AKI, and were Medicare-entitled for !2 yr before discharge. In this cohort, 3.1% survived todischarge with a diagnosis of AKI, and 5.3 per 1000 developed ESRD. Among patients who receivedtreatment for ESRD, 25.2% had a previous history of AKI. After adjustment for age, gender, race,diabetes, and hypertension, the hazard ratio for developing ESRD was 41.2 (95% confidence interval [CI]34.6 to 49.1) for patients with AKI and CKD relative to those without kidney disease, 13.0 (95% CI 10.6to 16.0) for patients with AKI and without previous CKD, and 8.4 (95% CI 7.4 to 9.6) for patients with CKDand without AKI. In summary, elderly individuals with AKI, particularly those with previously diagnosedCKD, are at significantly increased risk for ESRD, suggesting that episodes of AKI may accelerateprogression of renal disease.

J Am Soc Nephrol 20: 223–228, 2009. doi: 10.1681/ASN.2007080837

Although short-term consequences of acute kidneyinjury (AKI) have been extensively studied,1–4 the rateof development of end-stage renal disease (ESRD) af-ter AKI has been poorly defined in a representativesample. Moreover, the potential linkage between pa-tients with AKI, chronic kidney disease (CKD), andESRD has been poorly studied and remains ill defined,particularly among elderly individuals, who representthe fastest growing segment of the ESRD population.Since 1972, all patients who have ESRD and are eligi-ble for Social Security in the United States have beenentitled to all Medicare benefits, regardless of age. AllUS renal dialysis units and transplant facilities are re-quired to complete the Centers for Medicare & Med-icaid Services (CMS) End-Stage Renal Disease Medi-

cal Evidence Report (CMS-2728) for each patientreceiving initial treatment at that site. The Medical Ev-idence Report registers all patients who have ESRD

Received August 1, 2007. Accepted August 5, 2008.

Published online ahead of print. Publication date available atwww.jasn.org.

The interpretation and reporting of these data are the responsi-bility of the authors and in no way should be seen as an officialpolicy or interpretation of the US government.

Correspondence: Dr. Allan J. Collins, United States Renal DataSystem, 914 South 8th Street, Suite S-406, Minneapolis, MN55404. Phone: 612-347-5811; Fax: 612-347-5878; E-mail:acollins@usrds.org

Copyright ! 2009 by the American Society of Nephrology

CLINICAL RESEARCH www.jasn.org

J Am Soc Nephrol 20: 223–228, 2009 ISSN : 1046-6673/2001-223 223

with the US Renal Data System (USRDS), which maintains thedata. Thus, patients entitled to ESRD therapy after AKI can befound in the USRDS database.

The primary objective of our study was to determine thepostdischarge incidence rates and hazard ratios (HR) for de-veloping ESRD in elderly individuals who sustain AKI. A sec-ondary objective was to assess the difference in developingESRD between elderly individuals with and without CKD.

RESULTS

The incidence of AKI was 3.1% and baseline CKD was presentin 12.0% of the cohort in 2000 (Table 1). Among patients withAKI, the proportion with baseline CKD was 34.3%. Of patients

who survived their index hospitalizations, those with AKI andCKD had higher 2-yr mortality than those with AKI alone (Ta-ble 2). This may represent survivor bias; of the patients whohad AKI and died during the hospitalization, 64% had AKIalone, and 36% had AKI and CKD (data not shown).

The likelihood of initiating ESRD treatment among patientswith AKI increased steadily after discharge (Figure 1). Consid-ering patients with AKI only (Figure 1, top), the likelihood ofinitiating ESRD treatment after AKI was 0.96% within 30 d,2.69% within 180 d, 4.08% within 365 d, and 6.96% at the endof 2 yr of follow-up. The corresponding likelihoods for patientswithout AKI were 0.04, 0.14, 0.25, and 0.49%, respectively.Considering patients with both AKI and CKD (Figure 1, bot-tom), the likelihood of initiating ESRD treatment after AKIwas 1.61% within 30 d, 4.76% within 180 d, 7.91% within365 d, and 14.29% at the end of 2 yr of follow-up. Correspondinglikelihoods were 0.63, 1.68, 2.28, and 3.68% for those with AKIonly; 0.26, 0.86, 1.43, and 2.54% for those with CKD only; and0.01, 0.05, 0.11, and 0.25% for those with neither AKI nor CKD.

Table 3 shows the number of ESRD cases and HR for devel-opment of ESRD by patient demographics and clinical condi-tions. When AKI and CKD were considered independently,patients with AKI were 6.74 times more likely to develop ESRDthan those without injury, after adjustment for age, gender,race, diabetes, hypertension, and CKD. When AKI and CKDwere considered simultaneously, patients with AKI only (HR13.00) were approximately 54% more likely to develop ESRDthan those with CKD only (HR 8.43). Patients who had CKDand sustained AKI showed a striking multiplicative effect of theinteraction between the two conditions on the development ofESRD (HR 41.19), compared with patients with neither diag-nosis.

In patients who had CKD and had AKI coded as the princi-pal diagnosis, the rate of developing ESRD was 101.5 per 1000patients (Figure 2). After adjustment for age, gender, race, di-

Table 1. Patient characteristics: Medicare beneficiariesdischarged alive from hospitals, 2000

Characteristic Value

N 233,803Age (yr; mean) 79.2Men (%) 38.8Race (%)

white 89.0black 7.7other 3.3

Baseline condition (%)diabetes 27.2hypertension 64.9heart disease 69.3CKD 12.0

AKI (%) 3.1prior CKD 34.3

ESRD per 1000 patients 5.3discharged with AKI (%) 25.2

2-yr mortality (%) 29.1

Table 2. Incidence of AKI overall and among patients with previous CKD: Medicare patients discharged alive fromhospitals, 2000

CharacteristicAKI AKI and CKD

AKI No AKI AKI and CKD AKI Only CKD Only No AKI or CKD

n 7197 226,606 2467 4730 25,653 200,953Age (yr; mean) 80.1 79.2 79.9 80.2 79.3 79.1Men (%) 47.6 38.5 52.0 45.3 47.1 37.4Race (%)

white 83.6 89.2 81.8 84.5 86.7 89.5black 12.5 7.5 13.9 11.7 9.5 7.3other 4.0 3.3 4.3 3.8 3.8 3.2

Baseline condition (%)diabetes 41.6 26.8 50.3 37.0 37.7 25.4hypertension 78.0 64.5 84.9 74.3 77.3 62.9heart disease 84.1 68.9 91.1 80.5 83.0 67.1

Mortalityrate (%) 57.7 28.2 64.3 54.3 37.9 26.9HR (95% CI)a 2.38 (2.31 to 2.46) 1.00 3.24 (3.08 to 3.40) 2.48 (2.38 to 2.58) 1.45 (1.42 to 1.48) 1.00

aObtained using Cox proportional models with adjustment for age, gender, race, diabetes, and heart disease (CKD was included in the model for AKI).

CLINICAL RESEARCH www.jasn.org

224 Journal of the American Society of Nephrology J Am Soc Nephrol 20: 223–228, 2009

abetes, and hypertension, these patients were 55.36 times morelikely to develop ESRD than those with neither condition. Riskfor ESRD in patients with AKI coded as the principal diagnosisand without CKD (HR 15.04) was approximately 42% of therisk of those with both AKI coded as a secondary diagnosis andCKD (HR 36.24).

DISCUSSION

This study demonstrates for the first time the impact of AKI onthe hazard for developing ESRD in a representative sample ofMedicare beneficiaries. Initiation of ESRD treatment increasedsteadily after AKI occurred. Patients with both AKI and preex-isting or concurrent CKD showed a multiplicative effect onrisk for the development of ESRD.

AKI has generally been considered self-limiting, with agood prognosis when recovery is noted in hospital.5 Resultsfrom our study suggest that among patients who had AKIand did not have recognized preexisting CKD (n ! 4730),CKD was documented within 2 yr of the AKI occurrence in72.1%. These findings suggest that CKD may be unrecog-nized or that AKI can initiate CKD. Several early studies inwhich renal clearance measures were carefully performeddemonstrate that survivors of AKI may experience consid-erable decline in kidney function.6 – 8 Results from severalmore recent studies support these findings.9,10 A greaterlikelihood of initiating ESRD treatment among patientswho sustain AKI has also been noted in single-center studiesin which 8 to 10% of patients with AKI received long-termdialysis ("90 d).10 –15 Recently, two studies using the Coop-

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0 100 200 300 400 500 600 700

Pro

babi

lity

of E

SR

D

No AKIAKI

Log-rank test:P < 0.0001 DF=1

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0 100 200 300 400 500 600 700Days from hospital discharge

Pro

babi

lity

of E

SR

DNo AKI or CKDCKD onlyAKI onlyAKI and CKD

Log-rank test:P < 0.0001 DF=3

Figure 1. Estimated probability of initiating treatment of ESRD,using the Kaplan-Meier method. (Top) Curves by AKI status.(Bottom) Curves by AKI and CKD status. DF, degrees of freedom.

101.5

27.5

71.2

23.7 2.119.90

20

40

60

80

100

AKIP-CKD AKIP only AKIS-CKDAKIS only CKD only None

Cas

es p

er 1

000

pers

ons

55.36

15.04

36.24

12.43 8.43 1.000

10

20

30

40

50

60

70

AKIP-CKDAKIP only

AKIS-CKDAKIS only

CKD onlyNone

Haz

ard

ratio

Figure 2. Number of cases per 1000 persons (top) and HR(bottom) for ESRD by AKI and CKD status. AKIP, AKI codedas principal diagnosis; AKIS, AKI coded as secondary diag-nosis.

Table 3. Risk for ESRD in 2 yr after live hospitaldischarge, Medicare patients hospitalized in 2000

CharacteristicCases per

1000 PatientsHR (95% CI)a

Age (yr)67 to 70 8.00 1.0071 to 75 6.90 0.87 (0.74 to 1.02)76 to 80 5.65 0.72 (0.61 to 0.85)81 to 85 4.27 0.63 (0.52 to 0.76)!86 1.93 0.36 (0.28 to 0.46)

Gendermale 6.57 1.24 (1.11 to 1.39)female 4.50 1.00

Racewhite 4.35 1.00black 14.23 2.23 (1.93 to 2.56)other 10.13 1.66 (1.31 to 2.09)

Baseline diabetesyes 11.48 2.24 (1.99 to 2.52)no 2.99 1.00

Baseline hypertensionyes 7.07 1.91 (1.62 to 2.27)no 2.04 1.00

Baseline CKDyes 25.11 6.82 (6.05 to 7.67)no 2.60 1.00

AKIyes 43.35 6.74 (5.90 to 7.71)no 4.10 1.00

AKI and CKDboth AKI and CKD 79.45 41.19 (34.58 to 49.08)AKI only 24.52 13.00 (10.57 to 15.99)CKD only 19.88 8.43 (7.39 to 9.61)no AKI or CKD 2.08 1.00

aObtained using Cox proportional models with adjustment for age, gender,race, diabetes, and hypertension.

CLINICAL RESEARCHwww.jasn.org

J Am Soc Nephrol 20: 223–228, 2009 ESRD and Acute Kidney Injury 225

abetes, and hypertension, these patients were 55.36 times morelikely to develop ESRD than those with neither condition. Riskfor ESRD in patients with AKI coded as the principal diagnosisand without CKD (HR 15.04) was approximately 42% of therisk of those with both AKI coded as a secondary diagnosis andCKD (HR 36.24).

DISCUSSION

This study demonstrates for the first time the impact of AKI onthe hazard for developing ESRD in a representative sample ofMedicare beneficiaries. Initiation of ESRD treatment increasedsteadily after AKI occurred. Patients with both AKI and preex-isting or concurrent CKD showed a multiplicative effect onrisk for the development of ESRD.

AKI has generally been considered self-limiting, with agood prognosis when recovery is noted in hospital.5 Resultsfrom our study suggest that among patients who had AKIand did not have recognized preexisting CKD (n ! 4730),CKD was documented within 2 yr of the AKI occurrence in72.1%. These findings suggest that CKD may be unrecog-nized or that AKI can initiate CKD. Several early studies inwhich renal clearance measures were carefully performeddemonstrate that survivors of AKI may experience consid-erable decline in kidney function.6 – 8 Results from severalmore recent studies support these findings.9,10 A greaterlikelihood of initiating ESRD treatment among patientswho sustain AKI has also been noted in single-center studiesin which 8 to 10% of patients with AKI received long-termdialysis ("90 d).10 –15 Recently, two studies using the Coop-

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0 100 200 300 400 500 600 700

Pro

babi

lity

of E

SR

D

No AKIAKI

Log-rank test:P < 0.0001 DF=1

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0 100 200 300 400 500 600 700Days from hospital discharge

Pro

babi

lity

of E

SR

D

No AKI or CKDCKD onlyAKI onlyAKI and CKD

Log-rank test:P < 0.0001 DF=3

Figure 1. Estimated probability of initiating treatment of ESRD,using the Kaplan-Meier method. (Top) Curves by AKI status.(Bottom) Curves by AKI and CKD status. DF, degrees of freedom.

101.5

27.5

71.2

23.7 2.119.90

20

40

60

80

100

AKIP-CKD AKIP only AKIS-CKDAKIS only CKD only None

Cas

es p

er 1

000

pers

ons

55.36

15.04

36.24

12.43 8.43 1.000

10

20

30

40

50

60

70

AKIP-CKDAKIP only

AKIS-CKDAKIS only

CKD onlyNone

Haz

ard

ratio

Figure 2. Number of cases per 1000 persons (top) and HR(bottom) for ESRD by AKI and CKD status. AKIP, AKI codedas principal diagnosis; AKIS, AKI coded as secondary diag-nosis.

Table 3. Risk for ESRD in 2 yr after live hospitaldischarge, Medicare patients hospitalized in 2000

CharacteristicCases per

1000 PatientsHR (95% CI)a

Age (yr)67 to 70 8.00 1.0071 to 75 6.90 0.87 (0.74 to 1.02)76 to 80 5.65 0.72 (0.61 to 0.85)81 to 85 4.27 0.63 (0.52 to 0.76)!86 1.93 0.36 (0.28 to 0.46)

Gendermale 6.57 1.24 (1.11 to 1.39)female 4.50 1.00

Racewhite 4.35 1.00black 14.23 2.23 (1.93 to 2.56)other 10.13 1.66 (1.31 to 2.09)

Baseline diabetesyes 11.48 2.24 (1.99 to 2.52)no 2.99 1.00

Baseline hypertensionyes 7.07 1.91 (1.62 to 2.27)no 2.04 1.00

Baseline CKDyes 25.11 6.82 (6.05 to 7.67)no 2.60 1.00

AKIyes 43.35 6.74 (5.90 to 7.71)no 4.10 1.00

AKI and CKDboth AKI and CKD 79.45 41.19 (34.58 to 49.08)AKI only 24.52 13.00 (10.57 to 15.99)CKD only 19.88 8.43 (7.39 to 9.61)no AKI or CKD 2.08 1.00

aObtained using Cox proportional models with adjustment for age, gender,race, diabetes, and hypertension.

CLINICAL RESEARCHwww.jasn.org

J Am Soc Nephrol 20: 223–228, 2009 ESRD and Acute Kidney Injury 225

ORIGINAL CONTRIBUTION

Chronic Dialysis and Death Among Survivorsof Acute Kidney Injury Requiring DialysisRon Wald, MDCM, MPH, FRCPCRobert R. Quinn, MD, FRCPCJin Luo, MDPing Li, PhDDamon C. Scales, MD, PhD, FRCPCMuhammad M. Mamdani, PharmD,MPHJoel G. Ray, MD, MSc, FRCPCfor the University of Toronto AcuteKidney Injury Research Group

ACUTE KIDNEY INJURY, WHICHleads to a sudden decline inkidney function, is a com-mon and serious complica-tion of hospitalization in the adultpopulation.1-4 Many patients with se-vere acute kidney injury require initia-tion of hemodialysis or hemofiltration(termed dialysis throughout), and theirin-hospital mortality rate ranges from45% to 70%.5-12 Among those who sur-vive, as many as 15% require dialysisat the time of discharge.6,9

Little is known about the long-termoutcomes of patients with acute kid-ney injury requiring in-hospital dialy-sis, especially once they leave the hos-pital and recover enough kidneyfunction to be free of dialysis in theshort term. Prior studies were con-ducted in single centers or within re-stricted geographic regions,5,8,10,13,14 andfew studies observed patients for morethan 1 year.8,14,15 Accordingly, little isknown about a common clinical sce-nario in which prognostic implica-tions include resource allocation, spe-cialist follow-up, and dialysis planning.

The universal health care system inthe Canadian province of Ontario pro-vides a unique opportunity to studylong-term outcomes and health care useacross multiple study centers andamong an entire population. We evalu-ated the long-term risk of chronic di-alysis and death among hospitalized pa-tients who sustained acute kidneyinjury requiring acute dialysis.

METHODSDesign and ParticipantsWe completed a retrospective matchedcohort study using linked administra-tive health databases for all of On-tario. We identified all adults aged 19

years and older admitted to an acutecare hospital between July 1, 1996, andDecember 31, 2006, for whom length

For editorial comment see p 1227.

Author Affiliations: Divisions of Nephrology (Dr Wald)and General Internal Medicine (Dr Ray), St Michael’sHospital; The Keenan Research Centre, Li Ka ShingKnowledge Institute of St Michael’s Hospital (Drs Wald,Mamdani, and Ray); Department of Medicine, Uni-versity of Toronto (Drs Wald, Scales, Mamdani, andRay); Division of Nephrology (Dr Quinn) and Depart-ment of Critical Care Medicine (Dr Scales), Sunny-brook Health Sciences Centre; and the Institute forClinical Evaluative Sciences (Drs Luo, Li, Scales, Mam-dani, and Ray), all in Toronto, Ontario, Canada, andFaculty of Medicine, King Saud University, Riyadh,Saudi Arabia (Dr Mamdani).Membersof theUniversityofTorontoAcuteKidney In-jury Research Group are listed at the end of this article.Corresponding Author: Ron Wald, MDCM, MPH, FR-CPC, Division of Nephrology, St Michael’s Hospital,61 Queen St East, 9-140, Toronto, ON M5C 2T2,Canada (waldr@smh.toronto.on.ca).

Context Severe acute kidney injury among hospitalized patients often necessitatesinitiation of short-term dialysis. Little is known about the long-term outcome of thosewho survive to hospital discharge.

Objective To assess the risk of chronic dialysis and all-cause mortality in individualswho experience an episode of acute kidney injury requiring dialysis.

Design, Setting, and Participants We conducted a population-based cohort studyof all adult patients in Ontario, Canada, with acute kidney injury who required in-hospital dialysis and survived free of dialysis for at least 30 days after discharge betweenJuly 1, 1996, and December 31, 2006. These individuals were matched with patients with-out acute kidney injury or dialysis during their index hospitalization. Matching was byage plus or minus 5 years, sex, history of chronic kidney disease, receipt of mechanicalventilation during the index hospitalization, and a propensity score for developing acutekidney injury requiring dialysis. Patients were followed up until March 31, 2007.

Main Outcome Measures The primary end point was the need for chronic dialy-sis and the secondary outcome was all-cause mortality.

Results We identified 3769 adults with acute kidney injury requiring in-hospital di-alysis and 13 598 matched controls. The mean age was 62 years and median fol-low-up was 3 years. The incidence rate of chronic dialysis was 2.63 per 100 person-years among individuals with acute kidney injury requiring dialysis, and 0.91 per 100person-years among control participants (adjusted hazard ratio, 3.23; 95% confi-dence interval, 2.70-3.86). All-cause mortality rates were 10.10 and 10.85 per 100person-years, respectively (adjusted hazard ratio, 0.95; 95% confidence interval, 0.89-1.02).

Conclusions Acute kidney injury necessitating in-hospital dialysis was associated withan increased risk of chronic dialysis but not all-cause mortality.JAMA. 2009;302(11):1179-1185 www.jama.com

©2009 American Medical Association. All rights reserved. (Reprinted) JAMA, September 16, 2009—Vol 302, No. 11 1179

Downloaded From: http://jama.jamanetwork.com/ by a University of St. Andrews Library User on 05/18/2015

ORIGINAL CONTRIBUTION

Chronic Dialysis and Death Among Survivorsof Acute Kidney Injury Requiring DialysisRon Wald, MDCM, MPH, FRCPCRobert R. Quinn, MD, FRCPCJin Luo, MDPing Li, PhDDamon C. Scales, MD, PhD, FRCPCMuhammad M. Mamdani, PharmD,MPHJoel G. Ray, MD, MSc, FRCPCfor the University of Toronto AcuteKidney Injury Research Group

ACUTE KIDNEY INJURY, WHICHleads to a sudden decline inkidney function, is a com-mon and serious complica-tion of hospitalization in the adultpopulation.1-4 Many patients with se-vere acute kidney injury require initia-tion of hemodialysis or hemofiltration(termed dialysis throughout), and theirin-hospital mortality rate ranges from45% to 70%.5-12 Among those who sur-vive, as many as 15% require dialysisat the time of discharge.6,9

Little is known about the long-termoutcomes of patients with acute kid-ney injury requiring in-hospital dialy-sis, especially once they leave the hos-pital and recover enough kidneyfunction to be free of dialysis in theshort term. Prior studies were con-ducted in single centers or within re-stricted geographic regions,5,8,10,13,14 andfew studies observed patients for morethan 1 year.8,14,15 Accordingly, little isknown about a common clinical sce-nario in which prognostic implica-tions include resource allocation, spe-cialist follow-up, and dialysis planning.

The universal health care system inthe Canadian province of Ontario pro-vides a unique opportunity to studylong-term outcomes and health care useacross multiple study centers andamong an entire population. We evalu-ated the long-term risk of chronic di-alysis and death among hospitalized pa-tients who sustained acute kidneyinjury requiring acute dialysis.

METHODSDesign and ParticipantsWe completed a retrospective matchedcohort study using linked administra-tive health databases for all of On-tario. We identified all adults aged 19

years and older admitted to an acutecare hospital between July 1, 1996, andDecember 31, 2006, for whom length

For editorial comment see p 1227.

Author Affiliations: Divisions of Nephrology (Dr Wald)and General Internal Medicine (Dr Ray), St Michael’sHospital; The Keenan Research Centre, Li Ka ShingKnowledge Institute of St Michael’s Hospital (Drs Wald,Mamdani, and Ray); Department of Medicine, Uni-versity of Toronto (Drs Wald, Scales, Mamdani, andRay); Division of Nephrology (Dr Quinn) and Depart-ment of Critical Care Medicine (Dr Scales), Sunny-brook Health Sciences Centre; and the Institute forClinical Evaluative Sciences (Drs Luo, Li, Scales, Mam-dani, and Ray), all in Toronto, Ontario, Canada, andFaculty of Medicine, King Saud University, Riyadh,Saudi Arabia (Dr Mamdani).Membersof theUniversityofTorontoAcuteKidney In-jury Research Group are listed at the end of this article.Corresponding Author: Ron Wald, MDCM, MPH, FR-CPC, Division of Nephrology, St Michael’s Hospital,61 Queen St East, 9-140, Toronto, ON M5C 2T2,Canada (waldr@smh.toronto.on.ca).

Context Severe acute kidney injury among hospitalized patients often necessitatesinitiation of short-term dialysis. Little is known about the long-term outcome of thosewho survive to hospital discharge.

Objective To assess the risk of chronic dialysis and all-cause mortality in individualswho experience an episode of acute kidney injury requiring dialysis.

Design, Setting, and Participants We conducted a population-based cohort studyof all adult patients in Ontario, Canada, with acute kidney injury who required in-hospital dialysis and survived free of dialysis for at least 30 days after discharge betweenJuly 1, 1996, and December 31, 2006. These individuals were matched with patients with-out acute kidney injury or dialysis during their index hospitalization. Matching was byage plus or minus 5 years, sex, history of chronic kidney disease, receipt of mechanicalventilation during the index hospitalization, and a propensity score for developing acutekidney injury requiring dialysis. Patients were followed up until March 31, 2007.

Main Outcome Measures The primary end point was the need for chronic dialy-sis and the secondary outcome was all-cause mortality.

Results We identified 3769 adults with acute kidney injury requiring in-hospital di-alysis and 13 598 matched controls. The mean age was 62 years and median fol-low-up was 3 years. The incidence rate of chronic dialysis was 2.63 per 100 person-years among individuals with acute kidney injury requiring dialysis, and 0.91 per 100person-years among control participants (adjusted hazard ratio, 3.23; 95% confi-dence interval, 2.70-3.86). All-cause mortality rates were 10.10 and 10.85 per 100person-years, respectively (adjusted hazard ratio, 0.95; 95% confidence interval, 0.89-1.02).

Conclusions Acute kidney injury necessitating in-hospital dialysis was associated withan increased risk of chronic dialysis but not all-cause mortality.JAMA. 2009;302(11):1179-1185 www.jama.com

©2009 American Medical Association. All rights reserved. (Reprinted) JAMA, September 16, 2009—Vol 302, No. 11 1179

Downloaded From: http://jama.jamanetwork.com/ by a University of St. Andrews Library User on 05/18/2015

tional hazards models, and matched in-dividuals without acute kidney injuryand dialysis were the reference group.We further adjusted for age (continu-ous in years) and the propensity scorefor acute kidney injury and dialysis inthe multivariable models, in case therewas residual confounding beyond thematching process. Survival curves forchronic dialysis and all-cause mortal-ity were generated from unadjusted Coxmodels. Adjusted HRs for receipt ofchronic dialysis were further stratifiedaccording to participant characteristics.

Based on the actual number of par-ticipants enrolled in the current study,we had greater than 95% statisticalpower to detect an HR of 2.00 or morefor the primary study outcome at a con-ventional P value of .05. All P valueswere 2-sided, and the significance levelwas set at .05. Analyses were per-formed using SAS 9.1 for UNIX (SASInstitute Inc, Cary, North Carolina).

RESULTSWe identified 4066 individuals withacute kidney injury requiring dialysiswho survived at least 30 days after hos-

pital discharge without receiving di-alysis or further hospitalization(FIGURE 1). At least 1 match was iden-tified for 3769 (92.7%) of the exposedpatients (TABLE 1). Hence, 3769 indi-viduals with acute kidney injury and di-alysis and 13 598 matches with nei-ther acute kidney injury nor dialysiswere included in this study.

The mean age of the enrolled par-ticipants was 62 years, approximately40% were women, and the mean Charl-son Comorbidity Index score was 2.7(Table 1). Slightly more than 25% ofparticipants in each group had docu-mented chronic kidney disease in the5-year period before the index hospi-tal admission. During their hospital-ization, 47% of patients with acute kid-ney injury and dialysis receivedmechanical ventilation, compared with42% of those without acute kidney in-jury and dialysis. The respective ratesfor cardiac surgery were 11.5% and10.0%. Approximately 15% of partici-pants in each group had sepsis. Therates of cardiac and noncardiac arteri-ography were also comparable be-tween groups (Table 1).

Figure 1. Creation of the Acute KidneyInjury and Dialysis Study Cohort

4066 Survived >30 d postdischargewithout receiving dialysis orfurther hospitalization

3769 Included in analysis

8855 Survived to discharge

15 028 With a first hospitalization foracute kidney injury and dialysis

18 551 Patients hospitalized with acutekidney injury and dialysis July1996-December 2006

4789 Excluded4783 Had exclusionary event

within 30 d postdischarge3481 Received dialysis

915 Readmitted to hospital387 Died

6 Complete dataunavailable

3523 Excluded3321 Had exclusionary event in

prior 5 y

202 Hospital length of stay >180 d

2060 Previous acute kidneyinjury

1258 Received dialysis3 Kidney transplant

6173 Died during index hospitalization

297 Excluded (no matches were found)

Table 1. Characteristics of Hospitalized Patients by Group

Characteristic

With Acute Kidney Injuryand Dialysis at Index

Hospitalization (n = 3769)a

Without Acute Kidney Injuryor Dialysis at Index

Hospitalization (n = 13 598)a

Age at cohort entry, mean (SD), y 61.9 (16.3) 62.2 (16.3)Women 1504 (39.9) 5476 (40.3)Charlson Comorbidity Index, mean (SD), scoreb 2.7 (2.3) 2.8 (2.5)Chronic kidney disease !5 y before the index hospitalization admission date 1059 (28.1) 3684 (27.1)Comorbid conditions !5 y before the index hospitalization discharge date

Myocardial infarction 1692 (44.9) 5947 (43.7)Cancer 1566 (41.5) 5775 (42.5)Diabetes mellitus 1548 (41.1) 5741 (42.2)Heart failure 1437 (38.1) 5168 (38.0)Cerebrovascular disease 682 (18.1) 2515 (18.5)Liver disease 510 (13.5) 2019 (14.8)Peripheral vascular disease 147 (3.9) 513 (3.8)

Procedure or condition during index hospitalizationMechanical ventilation 1761 (46.7) 5700 (41.9)Sepsis 579 (15.4) 2045 (15.0)Cardiac surgery 435 (11.5) 1365 (10.0)Noncardiac arterial angiography 312 (8.3) 1210 (8.9)Coronary angiography with or without percutaneous coronary intervention 289 (7.7) 1153 (8.5)Abdominal aortic aneurysm repair 153 (4.1) 458 (3.4)

Propensity score for receipt of acute dialysis for acute kidney injury, mean (SD) 3.8 (2.6) 4.2 (2.3)Follow-up, median (interquartile range), y 3 (1-5) 3 (1-5)aAll dates are presented as No. (%) unless otherwise indicated.bCharlson Comorbidity Index scores range from 0 to 37, with a higher value indicating greater comorbidity.

CHRONIC DIALYSIS AND DEATH IN ACUTE KIDNEY INJURY SURVIVORS

©2009 American Medical Association. All rights reserved. (Reprinted) JAMA, September 16, 2009—Vol 302, No. 11 1181

Downloaded From: http://jama.jamanetwork.com/ by a University of St. Andrews Library User on 05/18/2015

Need for Chronic DialysisAfter a median (IQR) follow-up of 3(1-5) years, the incidence rates ofchronic dialysis were 2.63 and 0.91 per100 person-years among those with andwithout acute kidney injury and dialy-sis in the index hospitalization, respec-tively, corresponding to a crude HR of3.26 (95% CI, 2.73-3.89) (TABLE 2;FIGURE 2A). After adjusting for ageand the propensity score, the HRchanged little (3.23; 95% CI, 2.70-3.86; Table 2).

The increased risk of chronic di-alysis following a hospitalization com-plicated by acute kidney injury anddialysis was consistent across patientsubgroups (FIGURE 3). This was espe-cially so among patients without a pre-existing diagnosis of chronic kidneydisease (adjusted HR, 15.54; 95% CI,9.65-25.03).

All-Cause MortalityThe incidence rate for all-cause mortal-ity was 10.10 per 100 person-yearsamong patients with acute kidney in-jury and dialysis, which was slightlylower than among those without in-hospital acute kidney injury (Figure 2B).After adjusting for age and the propen-sity score, the future risk of death wasnot significantly different betweengroups (HR, 0.95; 95% CI, 0.89-1.02;Table 2).

Unmatched ParticipantsThere were 297 (7.3%) participantswith acute kidney injury and dialysiswho could not be matched to a suit-able control. Their respective inci-dence rates of chronic dialysis and all-cause mortality were 4.89 per 100person-years and 14.18 per 100 person-years, which were higher than rates for

the 3769 patients who were success-fully matched (Table 2).

COMMENTWe analyzed all patients admitted toOntario hospitals over a 10-year pe-riod and found that survivors of a hos-pitalization complicated by acute kid-ney injury requiring dialysis wereapproximately 3 times more likely torequire chronic dialysis compared withthose without acute kidney injury.However, no difference was observedbetween these groups for long-termmortality.

This study was designed to evaluatelong-term renal health among survi-vors of acute kidney injury receiving di-alysis in a large population with univer-sal access to health care. Patients withacute kidney injury have a high acuitylevel during their hospitalization and a

Table 2. Risk of Chronic Dialysis and All-Cause Mortality by Group

Outcome

Acute Kidney Injury and Dialysis atIndex Hospitalization (n = 3769)

Without Acute Kidney Injury or Dialysisat Index Hospitalization (n = 13 598)

Hazard Ratio(95% Confidence Interval)

No. of Events (%)

IncidenceRate Per 100Person-Years No. of Events (%)

IncidenceRate Per 100Person-Years Crudea Adjustedb

Chronic dialysis 322 (8.5) 2.63 403 (3.0) 0.91 3.26 (2.73-3.89) 3.23 (2.70-3.86)All-cause mortality 1311 (34.8) 10.10 4884 (35.9) 10.83 0.95 (0.88-1.02) 0.95 (0.89-1.02)aReflects the effect of acute kidney injury and dialysis vs matched individuals without acute kidney injury.bFurther adjusted for age (continuous in years) and the propensity score for acute kidney injury and dialysis.

Figure 2. Risk of Chronic Dialysis in Association With Acute Kidney Injury and Dialysis During Index Hospitalization

Acute kidney injury and dialysisNo acute kidney injury

20

15

5

10

0

No. at riskAcute kidney

injury and dialysisNo acute kidney

injury

3769

13 598

1Follow-up, y

Cum

ulat

ive

Ris

k of

Chr

onic

Dia

lysi

s, %

2 3 4 5 6 7 98 10

Chronic dialysis riskA

2761

10 224

2116

7850

1683

6080

1305

4639

964

3383

676

2342

462

1555

294

905

158

473

58

169

60

30

50

40

10

20

0

3769

13 598

1Follow-up, y

Cum

ulat

ive

Ris

k of

All-

Cau

se M

orta

lity,

%

2 3 4 5 6 7 98 10

Mortality riskB

2906

10 368

2262

8011

1805

6245

1404

4780

1038

3496

732

2429

495

1619

311

952

167

498

59

175

CHRONIC DIALYSIS AND DEATH IN ACUTE KIDNEY INJURY SURVIVORS

1182 JAMA, September 16, 2009—Vol 302, No. 11 (Reprinted) ©2009 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a University of St. Andrews Library User on 05/18/2015

ORIGINAL CONTRIBUTION

Chronic Dialysis and Death Among Survivorsof Acute Kidney Injury Requiring DialysisRon Wald, MDCM, MPH, FRCPCRobert R. Quinn, MD, FRCPCJin Luo, MDPing Li, PhDDamon C. Scales, MD, PhD, FRCPCMuhammad M. Mamdani, PharmD,MPHJoel G. Ray, MD, MSc, FRCPCfor the University of Toronto AcuteKidney Injury Research Group

ACUTE KIDNEY INJURY, WHICHleads to a sudden decline inkidney function, is a com-mon and serious complica-tion of hospitalization in the adultpopulation.1-4 Many patients with se-vere acute kidney injury require initia-tion of hemodialysis or hemofiltration(termed dialysis throughout), and theirin-hospital mortality rate ranges from45% to 70%.5-12 Among those who sur-vive, as many as 15% require dialysisat the time of discharge.6,9

Little is known about the long-termoutcomes of patients with acute kid-ney injury requiring in-hospital dialy-sis, especially once they leave the hos-pital and recover enough kidneyfunction to be free of dialysis in theshort term. Prior studies were con-ducted in single centers or within re-stricted geographic regions,5,8,10,13,14 andfew studies observed patients for morethan 1 year.8,14,15 Accordingly, little isknown about a common clinical sce-nario in which prognostic implica-tions include resource allocation, spe-cialist follow-up, and dialysis planning.

The universal health care system inthe Canadian province of Ontario pro-vides a unique opportunity to studylong-term outcomes and health care useacross multiple study centers andamong an entire population. We evalu-ated the long-term risk of chronic di-alysis and death among hospitalized pa-tients who sustained acute kidneyinjury requiring acute dialysis.

METHODSDesign and ParticipantsWe completed a retrospective matchedcohort study using linked administra-tive health databases for all of On-tario. We identified all adults aged 19

years and older admitted to an acutecare hospital between July 1, 1996, andDecember 31, 2006, for whom length

For editorial comment see p 1227.

Author Affiliations: Divisions of Nephrology (Dr Wald)and General Internal Medicine (Dr Ray), St Michael’sHospital; The Keenan Research Centre, Li Ka ShingKnowledge Institute of St Michael’s Hospital (Drs Wald,Mamdani, and Ray); Department of Medicine, Uni-versity of Toronto (Drs Wald, Scales, Mamdani, andRay); Division of Nephrology (Dr Quinn) and Depart-ment of Critical Care Medicine (Dr Scales), Sunny-brook Health Sciences Centre; and the Institute forClinical Evaluative Sciences (Drs Luo, Li, Scales, Mam-dani, and Ray), all in Toronto, Ontario, Canada